Journal of Muscle Research and Cell Motility 23: 367–375, 2002.
2003 Kluwer Academic Publishers. Printed in the Netherlands.
367
Mechanics and imaging of single DNA molecules
M. HEGNER* and W. GRANGE
Institute of Physics, NCCR Nanoscale Science, University of Basel, Klingelbergstrasse 82, CH-4056 Basel, Switzerland
Abstract
We review recent experiments that have revealed mechanical properties of single DNA molecules using advanced
manipulation and force sensing techniques (scanning force microscopy (SFM), optical or magnetic tweezers,
microneedles). From such measurements, intrinsic relevant parameters (persistence length, stretch modulus) as well
as their dependence on external parameters (non-physiological conditions, coating with binding agents or proteins)
are obtained on a single-molecule level. In addition, imaging of DNA molecules using SFM is presented.
Introduction
The study of nucleic-acid structure has experienced a
fundamental shift toward methods that are capable of
studying the physical properties of single molecules.
During the last 10 years several force-sensitive experiments have been used to stretch and manipulate single
DNA molecules. The application of hydrodynamic drag
(Bensimon et al., 1995; Perkins et al., 1997; Otobe and
Ohani, 2001), magnetic tweezers (Smith et al., 1992;
Strick et al., 2000a), glass needles (Leger et al., 1999),
optical tweezers (OT) (Smith et al., 1996; Bouchiat
et al., 1999; Hegner et al., 1999) and scanning force
microscopy (SFM) (Rief et al., 1999) allowed to investigate the mechanical response of single DNA molecules.
For all of these force-sensitive experiments, the following requirements have to be fulfilled. First, the molecule
has to be tethered in-between two microscopic objects
whose position can be precisely defined (nm accuracy).
Second, the force acting on moving parts has to be
quantified with force resolution in the picoNewton (pN)
regime. Force spectroscopy on single molecules is thus
technically most demanding. One can select a method of
generating tension that is tailored to a specific application for a specific range of forces. Today, the magnetic
beads and the OT technique are best suited to measure
the entropic and enthalpic force arising in singlemolecule DNA experiments. In particular, magnetic
tweezers are sensitive in the range of 0.01–10 pN and
OTs have sensitivity in a range of 0.1–200 pN (Grange
et al., 2002). The force limit (0.1 pN) of OT is two
orders of magnitudes lower than in conventional SFM
based techniques. The largest stretching forces (10–
10,000 pN) are generated by SFM, in which the molecule of interest is stretched between a surface and a
cantilever. For each of these methods the applied force
can be continuously and accurately varied; thus, the
energies of various nucleic acid and nucleic acid –
protein structural transitions can be quantified.
* To whom correspondence should be addressed
In this article we focus on recent experiments performed on single DNA molecules using the techniques
mentioned above. In particular, we discuss mechanical
properties and stress-induced transitions of a DNA
single molecule when stretched beyond its entropic
regime under both physiological and non-physiological
conditions.
This review is organized as follows. In the first part we
briefly mention simple theoretical models often used to
describe the behavior of a polymer either at low
(entropic regime) or high (enthalpic regime) force.
Anchoring the molecules is certainly a key point in
single-molecule measurements, which determines the
success of the experiment. This is discussed in details in a
second part. Then, we present typical force vs. extension
curves obtained either on dsDNA, ssDNA, and discuss
how external parameters (salt, surrounding proteins,
etc.) can affect the observed mechanical properties. We
also briefly mention experiments performed to twist,
unzip or unbind single DNA molecules. Finally, recent
measurements on DNA imaging are highlighted.
Models for DNA elasticity
The relevance of single-molecule force experiments lies
in the fact that important parameters that describe the
stiffness of the polymer (e.g. DNA) can be extracted.
Such parameters are not accessible when measurements
are performed on an ensemble, simply because this
would need the synchronization of all stretching events.
Again, we emphasize that such stretching experiments
(force vs. distance) need to have a very high-force
sensitivity. To give order of magnitudes, the characteristic force required to align a polymer made of identical
rigid units of length b is kT/b and is only about 0.1 pN
for dsDNA.
Entropic elasticity
Depending on the polymer different models are applied to correctly describe its entropic elastic behavior
368
(Merkel, 2001). The freely jointed chain (FJC) model
(Flory,1969) considers a chain of identical N rigid
subunits that are freely able to rotate (i.e. the angle
between the different segments is not fixed):
x
Fb
kB T
¼ coth
L
kB T
Fb
ð1Þ
where x is end-to-end length, L is the contour length of
the polymer (L¼Nb), and b denotes the Kuhn length.
Alternatively, the wormlike chain model (WLC) can be
used (Bustamante et al., 1994; Marko and Siggia, 1995;
Rivetti et al., 1998). In the WLC model, the polymer is
treated as an uniform flexible rod and bending costs
energy. In this case, the force-extension relation reads:
FA
1
x 2 x 1
1
¼
þ kB T 4
L
L 4
ð2Þ
where A is known as the persistence length (the distance
along which the molecule can be considered as rigid).
Note that (i) Equation (2) is only an approximation to
the WLC model and additional correction terms should
be added for a more rigorous treatment (Bouchiat et al.,
1999) (ii) the persistence length A is defined to be half of
the Kuhn length b.
Enthalpic elasticity
At higher forces (above 5–10 pN for DNA), the elastic
response of the polymer under an applied tension is not
more purely entropic. In this case, enthaplic contributions have to be taken into account (Odijk, 1995) and
experimental data are usually fitted using an extensible
WLC model (we consider that – due to some stretchability – the contour length linearly increases with the
applied force) (Smith et al., 1996):
x
1 kB T 1=2 F
¼1
þ
L
2 FA
S
ð3Þ
where S denotes the stretch modulus of the polymer.
Alternatively, models were the Kuhn segments are able
to stretched have been proposed (Smith et al., 1996).
suited due to its high strength. In SFM experiments dealing with short molecules, single thiol-modified ssDNA
molecules were covalently attached to an activated glass
surface (silanized with an aminopropyltriethoxysilane) through a poly(ethylene glycol)-a-maleimide-x-Nhydroxysuccinimide-ester spacer (Chrisey et al., 1996;
Strunz et al., 1999). Both the glass and the tip surfaces
were treated in parallel except that different ssDNA (i.e.
complementary ssDNA) were attached on the AFM tip
and the glass surface, respectively. The experiment then
just consists of approaching the AFM to the surface till
a bond (due to hybridization) forms between the
complementary ssDNA strands (Figure 1A., upper panel). Note that the use of spacers in this case (i.e. short
molecules) is of prime importance because this reduces
non-specific interactions. For OT experiments, Hegner
(Hegner, 2000) developed a procedure for covalent
binding of DNA molecules to polystyrene microspheres.
Briefly, it consists (i) to chemically activate microspheres
and covalently attach a thiol- or amino-modified 5-end
of a dsDNA to the surface of the bead (ii) to attach a
ligand (e.g. biotin) to the opposite 3-end. The ligand can
then interact with receptors immobilized on a second set
of beads, as shown in Figure 1B. (upper panel). This
procedure allows preparation of DNA-microspheres in
advance and storage of these beads for months. Moreover, high forces can be reached (200 pN) (Hegner
et al., 1999; Wuite et al., 2000). This is not possible
when using digoxygenin–antidigoxygenin recognition
(i.e. non-covalent coupling) (Leger et al., 1998; Merkel,
2001). Notice that a similar procedure was applied in
magnetic tweezers experiments (Haber and Wirtz, 2000).
Non-covalent coupling is the most widely used method in OT experiments, certainly due to the ease of
preparation. Many groups have worked on lambda
DNA or its linearized fragments, which can be labeled
at both its 5-ends with biotinylated nucleotides
(Figure 1B., lower panel) using nucleotide incorporation
with DNA polymerase enzymes. In contrast to covalent
coupling, both ends have the same bioreactive group
(biotin) and no beads stock solution can be prepared
(Davenport et al., 2000).
Finally, DNA can be immobilized between a surface
and a tip of an AFM cantilever using non-specific
interactions (Rief et al., 1999) (Figure 1A., lower panel).
Immobilization of molecules to surfaces
Overview on single-DNA experiments
A challenge in single-molecule experiments is to specifically and strongly attach molecules onto surfaces. In
this review, we focus on DNA single molecules although
the techniques presented in this section might be applied
to other biological samples. Different procedures exist
depending on the requirements whether (i) a covalent, a
non-covalent, or a non-specific coupling has to be
achieved (ii), or a SFM or an optical (magnetic) tweezers
experimental setup is used.
Covalent coupling – although it involves more steps
and is somehow more difficult to achieve – is always best
Mechanical properties of bare DNA under physiological
conditions
Low-force regime of dsDNA
Figure 2 shows a typical dsDNA force vs. fractional
extension (x/L) curve obtained with a modified linearized pTYB1 plasmid (7477 bp). For this experiment, a
20-bp thiol-labeled dsDNA linker was ligated to the 5end of the fragment and the single strand overhang of
the 3-end was biotinylated with DNA polymerase
369
Fig. 1. Possible experimental arrangements for stretching single dsDNA molecules using either a SFM (A) or OT (B) in liquid and room
temperature. A. Top panel: complementary ssDNA molecules are covalently attached to a treated AFM tip and a treated glass (Strunz et al.,
1999) or gold surface (Boland and Ratner, 1995). When approaching the tip to the surface, a bond may form between the two complementary
ssDNAs (DNA hybridization). Bottom panel: an untreated tip is pressed onto an untreated glass surface. Due to non-specific interactions,
dsDNA can be picked up and subsequently stretched (the tip has to be pressed to the surface and large forces (up to a few nN) have to be applied).
B. Top panel: a dsDNA molecule is covalently attached to a chemically modified polystyrene microsphere (bond A), which is held by the optical
trap. The free end of the dsDNA molecule has been labeled with a ligand that can interact with receptors immobilized on a sphere (held by suction
on a moveable micropipette) (bond B). This bond (B) can be a weak non-covalent bond (e.g. biotin–streptavidin) (Hegner, 2000). Bottom panel:
two microspheres are coated with identical receptors (e.g. streptavidin). One of the beads is held by the optical trap, while the second is held by
suction on a moveable micropipette. A dsDNA, which is modified at its opposite 5¢-ends with biotin ligands, is injected in the chamber and a
continuous flow is applied. Upon attachment of the DNA to the bead in the optical trap (bond A), the measured drag force on the bead increases.
The pipette is then moved in the vicinity of the optical trap to attach the free end of the DNA to the pipette bead (bond A). For this experiment,
continuous flow is required to prevent looping (Davenport et al., 2000).
Fig. 2. Conversion of dsDNA into ssDNA (OT experiment, pTYB1 plasmid). For dsDNA single molecules (squares), the force vs. fractional
extension (x/L) curve shows a typical overstretching plateau at 68 pN (150 mM NaCl). At low forces (smaller than 5–10 pN, entropic regime),
the mechanics of dsDNA can be well described in terms of an inextensible WLC model (Equation 2, dots). At higher forces (smaller than 68 pN),
enthalpic contributions have to be taken into account to correctly describe the observed mechanics [extensible WLC model (Equation 3, solid
line)]. dsDNA can be converted into ssDNA if a high force (about 140 pN) is applied (circles).
enzymes (Husale et al., 2002). Covalent coupling to
polystyrene microspheres was performed in a similar
procedure as described in the previous section (Hegner,
2000). At low forces [smaller than 10 pN (Odijk, 1995)],
the mechanics of the dsDNA (squares) follows a WLC
behavior [Equation (2), Figure 2 (dots)] with a persistence length A of 50 nm at 150 mM NaCl and pH 7.2
(Bustamante et al., 2000a; Strick et al., 2000b). Note
that for thick polymer where the sub-units are held
together by several bonds in parallel (e.g. dsDNA), the
FJC model fails to describe the observed mechanics
(Merkel, 2001). Although OT are widely used to
investigate stretching of single molecules at low force
(entropic regime), we would like to point out that
magnetic tweezers is always a technique of choice due to
its high force-resolution (see Introduction).
As described previously, an extensible WLC model
[Figure 2 (solid line)] provides an excellent agreement
370
for forces in the range of 10–68 pN. From a linear fit,
enthalpic parameters such as the stretch modulus can be
determined. Reported values at standard buffer conditions (150 mM NaCl, pH 7.2) are in the order of
1500 pN for bare dsDNA (Smith et al., 1996; Bustamante et al., 2000a).
Overstretching transition of bare dsDNA
At forces of about 68 pN (Cluzel et al., 1996; Smith
et al., 1996; Rief et al., 1999), a highly cooperative
transition can be seen: over a range of 2–3 pN, the
dsDNA is stretched to about 1.7 times its natural
contour length (B-DNA form). The occurrence of this
plateau, first seen with OT (Smith et al., 1996), is still a
subject of debate. Different explanations, either forceinduced melting (Rouzina and Bloomfield, 2001a, b) or
structural transition (S-DNA) (Lebrun and Lavery,
1996; Haijun et al., 1999) have been proposed to explain
the experimental observations. Due to (i) the high
cooperativity and (ii) the observed hysteresis while
relaxing the molecule, the overstretching plateau could
be caused through a force-induced melting (i.e. conversion of dsDNA into ssDNA). When subsequently
stretching the dsDNA apart its opposite 5–5 (Smith
et al., 1996) or its 5–3 ends (Hegner et al., 1999; Husale
et al., 2002), the molecular link should either fall apart
(5–5 pulling) or the force vs. extension curve should
show a behavior typical of ssDNA (5–3 pulling).
However, neither of these phenomena are observed in
experiments. Moreover, it was shown that partially
cross-linked DNA (5% intercalation of psoralen) only
slightly modified the overstretching transition (Smith
et al., 1996). For this reason, many groups have
suggested that B-DNA undergoes a structural transition
into a new form called S-DNA. To date, calculations
based on this assumption (i.e. DNA remains in a
dsDNA form during the overstretching transition) were
not able to predict the experimental trends (e.g. width of
the transition, or value of the overstretching plateau). In
contrast, models based only on purely thermodynamical
arguments found that melting of B-DNA should occur
at 60–80 pN of external applied force and correctly
estimated the temperature-, salt- and pH-dependence of
the overstretching plateau (Williams et al., 2001a, b;
Wenner et al., 2002).
Mechanical properties of bare ssDNA under physiological
conditions
Also shown in Figure 2 (circles) is the elastic property of
bare ssDNA (150 mM NaCl, pH 7.2). Note that ssDNA
was directly obtained from dsDNA by applying a high
tension (see arrow in Figure 2) on the single molecule
(Hegner et al., 1999). For single-chain polymer, it is
believed that a FJC model is applicable. This is
especially true at 150 mM salt, where ions of different
charges are perfectly counterbalanced (theta solvent, no
excluded-volume effects). Although the force vs. fractional extension curve can be fitted with a modified FJC
model [where the Kuhn segments are allowed to be
stretched (Smith et al., 1996)], it is necessary in this case
to shrink the expected contour length by 15%. In
contrast, a non-extensible WLC model (with appropriate contour length) qualitatively describes the mechanics
up to about 30 pN. Typical values reported for the
persistence length are in the order of 0.75 nm (Smith
et al., 1996; Rivetti et al., 1998).
AFM vs. OT and magnetic tweezers
Although AFM based techniques have been applied in
the past to investigate mechanical properties of single
polymers, intrinsic relevant parameters such as the
persistence length are not accessible to this technique,
mainly because of the large thermal noise of commercial
AFM cantilever (10 pN, in liquid and at room temperature). However, AFM is a technique of choice to reveal
overall mechanical trends, as shown for example by Rief
et al. (1999). For instance, these studies have carefully
studied DNA mechanics in the high-force regime and
have shown that (i) the overstretching transition represents thermodynamical equilibrium (independent of
pulling rate) and (ii) the melting of dsDNA into ssDNA
(observed at higher forces, see Figure 2) is a nonequilibrium process (depends on pulling rate).
Mechanical properties of DNA under non-physiological
conditions
When DNA is stretched under non-physiological conditions [change in solvent (Smith et al., 1996), salt
(Smith et al., 1992, 1996; Baumann et al., 1997; Wenner
et al., 2002), pH (Williams et al., 2001a) or temperature
(Williams et al., 2001b)], the mechanics is strongly
modified. Varying such external parameters affects both
the value of the overstretching force as well as the
persistence length. For instance, Smith et al. (1992)
found an increase of the persistence length A as well as a
decrease of the overstretching force when decreasing the
salt concentration.
Mechanical properties of coated DNA: binding agents,
proteins
We now investigate how mechanical properties are
affected when some agents (Ethidium Bromide (EtBr) or
SYBR Green I) intercalate/bind to dsDNA (Figure 3).
Although such measurements can be of fundamental
interest, such studies can also be of great use for
screening purposes without the need of competitive
assays (Husale et al., 2002). As expected, an extensible
WLC model still provides an excellent description of the
mechanical properties when dsDNA is exposed to such
binding ligands. For EtBr [1 lg/ml] (Figure 3, stars), we
find a decrease in persistence length by factor about two
(A ¼ 25 nm), a reduction in the stretch modulus of
about six (S ¼ 250 pN), and a change in contour length
of 25%, respectively. In addition, dsDNA does not
show in this case any cooperativity at high force. In
contrast (data not shown), SYBR Green I only slightly
371
Fig. 3. Comparison of the mechanical properties of (i) naked dsDNA (dots) (ii) dsDNA coated with EtBr (stars) (Husale et al., 2002) (ii) dsDNA
coated with RecA (circles) (150 mM NaCl). For this experiment, a pTYB1 dsDNA plasmid (7477 bp) was used. The change in persistence length
is well evidenced by the change in slope steepness when approaching the contour length. Coated dsDNA with EtBr (RecA) has a smaller (higher)
persistence length than naked dsDNA and therefore the force vs. extension curve shows a less steep (steeper) slope upon approaching the contour
length. Note the drastic increase of the contour length of bare dsDNA when coated with RecA.
affects the entropic elastic behavior of bare dsDNA
mechanics (A ¼ 40 nm). This would explain why –
among all fluorescence dyes – SYBR Green is best
suited when enzymatic reactions have to be performed
(Schäfer et al., 2000). Note, however, that the mechanics
are altered at higher forces since we measure a stretch
modulus of about 500 pN. Still, a small overstretching
plateau can be seen at 80 pN.
We emphasize that the differences in mechanics
obtained for these two latter compounds can be understood easily by the fact that – unlike EtBr – SYBR
Green I does not intercalate between stacked base pairs.
Rather, SYBR Green I is thought to bind in the minor
groove of DNA.
Although most Protein-DNA complexes affect the
mechanical properties on a local scale only [DNAenzymes (Bustamante and Rivetti, 1996; Bustamante
et al., 2000b), initiation protein complexes (e.g. tbp)],
some of them act on a larger scale (from a few tens of
bps to a few thousand). For instance, RecA is known to
lengthen and stiffen DNA chains (Figure 3, circles) to
facilitate the pairing of homologous sequences (Rocca
and Cox, 1997). Hegner et al. (1999) investigated the
polymerization of RecA on dsDNA and the mechanics
of both RecA-dsDNA and – ssDNA complexes. In
agreement with biochemical ensemble experiments, the
study of mechanical properties of RecA-dsDNA filaments showed that RecA binds to one DNA strand,
while the second strand does not adhere tightly to the
RecA helix and that this strand is able to slide past the
protein component.
Finally, studies have been performed on single
chromatin fibers to reveal how assembly and disassembly of chromatin takes place on dsDNA upon pulling
(Cuy and Bustamante, 2000; Bennick et al., 2001). At
low force (smaller than 10 pN), nucleosomes are able
to assemble and the fiber can be reversibly stretched
without disassemble the chromatin structure. Interestingly, the force vs. extension curves show discrete,
sudden drops at higher forces. This suggests an irreversible dissociation of the nucleosomal core particles
from the DNA.
Twisting dsDNA
Magnetic tweezers have the capability to easily and
accurately rotate magnetic handles (see Strick et al.,
2000b for a recent review). It makes it therefore possible
to study (i) torsionally constrained DNA single molecules [for instance the B to P-DNA transition where the
helical pitch is strongly reduced (Allemand et al., 1998)]
and (ii) the relaxation of supercoiled DNA by individual
enzymes (e.g. topoisomerase) (Strick et al., 2000c).
Unzipping and unbinding dsDNA
In the preceeding sections we have seen that stretching
or twisting a single DNA molecule yields important
information on either the mechanics (persistence length,
stretch modulus) or the stress-induced transitions
(overstretching transition, P-DNA form, etc.). Additional information can be obtained when unzipping
or unbinding dsDNA molecules, as briefly described
below.
Despite inherent experimental and theoretical limitations, unzipping experiments might have potential
applications for large-scale DNA sequencing (EssevazRoulet et al., 1997; Rief et al., 1999). An interesting
phenomenon observed in such experiments was the
occurrence of stick-slip events, similar to what can be
372
found on a macroscopic scale (Bockelmann et al.,
1997). In other words, strain energy accumulates while
stretching the dsDNA single molecule (the force increases) and at a certain force threshold (different for
A–T and C–G base pairs) the DNA unzipps in a cooperative manner. These experimental findings have
stimulated a number of theoretical investigations
(Lubensky and Nelson, 2000; Cocco et al., 2001; Bockelmann et al., 2002).
Closely related experiments to unzipping are unbinding experiments, where the force needed to fully separate
both strands is recorded. Unbinding forces of weak,
non-covalent bonds have been measured by SFM
(Dammer et al., 1996) or biomembrane force probes
(Evans and Ritchie, 1997). Initially, these SFM measurements focused on feasibility studies to measure
single biomolecular interactions. For instance, Boland
and Ratner (1995) have first tried to measure the force
necessary to break a single DNA base pair. But recently
a few groups showed that these single-molecule experiments give a direct link to ensemble experiments where
thermodynamic data are measured. Moreover, such
studies give insight into the geometry of the energy
landscape of a biomolecular bond (Evans, 1998; Strunz
et al., 2000; Evans, 2001; Merkel, 2001). An inherent
feature of these experiments is that unbinding forces
depend on the rate of loading and on the details of the
functional relationship between bond lifetime and the
applied force. These points have been addressed in
details for dsDNA: (i) loading rate and length (bp)
dependence (Strunz et al., 1999) and (ii) entropic contributions to the energy landscape (Schumakovitch
et al., 2002). An important result (in agreement with
thermodynamic data) was the finding – on the singlemolecule level – of cooperative unbinding of base pairs
in the DNA duplex.
DNA imaged by SFM
SFM is capable of generating images within ranges of
resolution that are of particular interest in biology.
Although true atomic resolution may not be possible
with biological samples at the moment, a great deal of
information can still be obtained from images that show
details at a slightly lower level of resolution (Bustamante
and Rivetti, 1996; Fotiadis et al., 2002). To date, only a
few scanning tunneling microscopy (STM) experiments have been reported on DNA (Guckenberger
et al., 1994; Tanaka et al., 1999). Since STM ope-ration
in native conditions (i.e. liquid) is difficult to achieve, we
believe that SFM is the technique of choice. For this
reason, STM experiments will not be discussed in this
review.
Imaging techniques
A decade ago, imaging of dsDNA under ambient
conditions was performed using contact mode imaging
were a cantilever tip is scanned across the surface while
applying a constant force. The advantage of SFM over
other techniques is that (i) direct imaging is possible
without staining of the molecule and (ii) a simple
deposition techniques onto flat surfaces can be applied.
Erie and colleagues (Erie et al., 1994) imaged lambda
Cro protein when bound as a single dimer or multiple
dimers to its three operator sites on dsDNA. They
observed that bending of the non-specific sites is
advantageous for a protein such as Cro that bends its
specific site, because it increases the binding specificity
of the protein. First images of dsDNA in liquids using
contact mode showed that technical improvements
were needed to allow higher resolution for imaging
dsDNA interacting with enzymes (Hegner et al., 1993)
(Figure 4).
Nowadays, the dynamic-mode operation is the method of choice to image DNA molecules (in air or in
liquids). Rivetti et al. (1999) analyzed the structure of
E. coli RNA polymerase-sigma open promoter complex.
The high-resolution capabilities of SFM allowed a
detailed analysis of a large number of molecules and
showed that the dsDNA contour length of open
promoter is reduced by approximately 30 nm (90 bp)
relative to the free dsDNA. The dsDNA bend angle
measured with different methods varied from 55 to 88.
This strongly supported the notion that during transcription initiation, the promoter DNA wraps nearly
300 around the polymerase.
Mechanics of DNA deposited on flat surfaces
An important point to be considered is the fact that for
imaging, the DNA is required to be deposited on a flat
surface and that this deposition can affect parameters
involving DNA mechanics. If proteins have to interact
with the DNA, the conditions for binding the DNA to a
surface might not be suitable for an optimal proteinDNA interaction.
Rivetti et al. (1996, 1998) developed suitable procedures to deposit DNA molecules onto freshly cleaved
mica. This method allows the DNA to equilibrate on the
surface as in an ideal two-dimensional solution. Under
equilibration conditions, this study indicated that the
SFM can be used to determine the persistence length of
DNA molecules to a high degree of precision. They
reexamined the DNA length measurement recently
and revealed a discrete, size-dependent, shortening of
DNA molecules deposited onto mica under low salt
conditions (Rivetti and Codeluppi, 2001). Awareness
of this structural alteration, which can be attributed to
a partial transition from B- to A-form DNA, may lead
to a more correct interpretation of DNA molecules
or protein-DNA complexes imaged by SFM in the
future.
We have shown previously that small agents can
considerably affect the mechanics. In a recent study,
Coury et al. (1996) presented a procedure to detect these
properties by measuring the contour length of the
373
Fig. 4. Tapping-mode SFM image of an enzyme (HindIII) interacting with a plasmid DNA (pNEB193). As suggested by the 5 nm gap visible
next to the enzyme, the deposition of the protein DNA complex took place immediately after the digestion of the closed plasmid. Measurements
were made under ambient conditions [deposition buffer 4 mM Hepes pH 7.2, 10 mM NaCl, 2 mM MgCl2 (Rivetti et al., 1996)].
dsDNA molecule using SFM. They incubated the bare
DNA molecules with the specific agent and subsequently
deposited the modified molecules onto a mica surface.
They investigated the amount of extension relative to
the contour length. It was shown that the fraction of
bound molecules could be estimated and an affinity
could be determined by subjecting the DNA to various
amounts of ligands. Such an approach reveals some of
these parameters, but has the drawback that molecules
have to be placed onto a surface in order to be accessible
to the SFM, which affects the binding of the ligand to
the DNA.
Protein-dsDNA interaction measured in physiological
buffer solutions
The group of Bustamante provided a series of images
with macromolecular resolution, which showed details
on the mechanisms by which RNA polymerase nonspecifically translocates along DNA (Guthold et al.,
1999). The dynamics of nonspecific and specific Escherichia coli RNA polymerase (RNAP)-DNA complexes
were directly observed using SFM operating in buffer
solution. Imaging conditions were found in which DNA
molecules were adsorbed onto mica strongly enough to
be imaged. Moreover, the same molecule is bound
loosely enough to be able to diffuse on the surface. In
sequential images of non-specific complexes, RNAP was
seen to slide along DNA, performing a one-dimensional
random walk. Note that in these studies the enzyme
remained fixed on the surface whereas the DNA was
sliding through (along) the polymerase enzyme. It was
observed that mica-bound transcription complexes
showed a transcription rate, which was about three
times smaller than the rate of complexes in solution.
This assay confirmed that the biological function of the
enzyme is considerably affected by the surface but that
‘native’ imaging is possible using SFM in liquids.
Mapping of dsDNA with SFM and time-lapse imaging
of protein-DNA interactions
Some other specific highlights in the literature include
the mapping of DNA using restriction enzymes (Allison
et al., 1996, 1997). These experiments showed that SFM
can be used to perform restriction mapping on individual cosmid clones. A mutant EcoRI endonuclease
was site-specifically bound to DNA. Distances between endonuclease molecules bound to lambda DNA
were measured and compared to known values. These
374
experiments demonstrate the accuracy of SFM mapping
to better than 1 kbp. These results may be extended to
identify other important site-specific protein-DNA interactions, such as transcription factor and mismatch
repair enzyme binding, which are difficult to resolve by
other techniques. In addition, time-lapsed microscopy
was used to observe dynamic interactions of proteins on
dsDNA. In these series of experiments, dynamic interactions of the tumor suppressor protein p53 with a
DNA fragment containing a p53-specific recognition
sequence were directly observed by time-lapse tappingmode SFM in liquid (Jiao et al., 2001). As in other
studies (Guthold et al., 1999) the divalent cation Mg2+
was used to loosely attach both DNA and p53 to a mica
surface so that they could be imaged by the SFM while
interacting with each other. Various interactions of p53
with DNA were observed, including dissociation/reassociation, sliding and possibly direct binding to the
specific sequence.
These authors visualized two modes of target recognition of p53: (a) direct binding, and (b) initial
non-specific binding with subsequent translocation by
one-dimensional diffusion of the protein along the DNA
to the specific site. It is to be foreseen that in the future,
i.e. when both higher resolution (Li et al., 1999) and
faster imaging in liquids (time to require one image
<1 min) will be routine, time lapsed-imaging of SFM in
liquids will allow to get insights of unprecedented
quality and dynamics.
Acknowledgements
Financial support of the NCCR ‘Nanoscale Science’ the
Swiss National Science Foundation and the ELTEM Regio Project Nanotechnology is gratefully acknowledged.
References
Allemand F, Bensimon D, Lavery R and Croquette V (1998) Stretched
and overwound DNA forms a Pauling-like structure with exposed
bases. Proc Natl Acad Sci USA 95: 14,152–14,157.
Allison DP, Kerper PS, Doktycz MJ, Spain JA, Modrich P, Larimer
FW, Thundat T and Warmack RJ (1996) Direct atomic force
microscope imaging of EcoRI endonuclease site specifically bound
to plasmid DNA molecules. Proc Natl Acad Sci USA 93: 8826–
8829.
Allison DP, Kerper PS, Doktycz MJ, Thundat T, Modrich P, Larimer
FW, Johnson DK, Hoyt PR, Mucenski ML and Warmack RJ
(1997) Mapping individual cosmid DNAs by direct AFM imaging.
Genomics 41: 379–384.
Baumann CG, Smith SB, Bloomfield VA and Bustamante C (1997)
Ionic effects on the elasticity of single DNA molecules. Proc Natl
Acad Sci USA 94: 6185–6190.
Bennink ML, Leuba SH, Leno, GH, Zlatanova, J, DE Grooth BJ and
Greve J (2001) Unfolding individual nucleosomes by stretching
single chromatin fibers with optical tweezers. Nat Struct Biol 8:
606–610.
Bensimon D, Simon AJ, Croquette V and Bensimon A (1995) Stretching DNA with receding meniscus: experiments and models, Phys
Rev Lett 74: 4754–4757.
Bockelmann U, Essevaz-Roulet B and Heslot M (1997) Molecular
stick-slip motion revealed by opening DNA with piconewton
forces. Phys Rev Lett 79: 4489–4492.
Bockelmann U, Thomen PH, Essevaz-Roulet B, Viasnoff V and Heslot
M (2002) Unzipping DNA with optical tweezers: high sensitivity
and force flips. Biophys J 82: 1537–1553.
Boland T and Ratner BD (1995) Direct measurement of hydrogen
bonding in DNA nucleotide bases by atomic force microscopy.
Proc Natl Acad Sci USA 92: 5297–5301.
Bouchiat C, Wang MD, Allemand JF, Strick T, Block SM and
Croquette V (1999) Estimating the persistence length of a wormlike chain molecule from force-extension measurements. Biophys J
76: 409–413.
Bustamante C, Marko JF, Siggia ED and Smith S (1994) Entropic Elasticity Of Lambda-Phage DNA. Science 265: 1599–
1600.
Bustamante C and Rivetti C (1996) Visualizing protein-nucleic acid
interactions on a large scale with the scanning force microscope.
Ann Rev Biophys Biomol Struct 25: 395–429.
Bustamante C, Smith SB, Liphardt J and Smith D (2000a) Singlemolecules studies of DNA mechanics. Curr Opi Struct Biol 10:
279–285.
Bustamante C, Macosko JC and Wuite JL (2000b) Grabbing the cat by
the tail: manipulating molecules one by one. Nat Rev Mol Cell Biol
1: 130–136.
Cluzel P, Lebrun A, Heller C, Lavery R, Viovy JL, Chatenay D and
Caron F (1996) DNA: an extensible molecule. Science 271: 792–
794.
Chrisey LA, Lee GU and O’Ferrall CE (1996) Covalent attachment of
synthetic DNA to self-assembled monolayer films. Nucleic Acids
Res 24: 3031–3039.
Cocco S, Monasson R and Marko JF (2001) Force and kinetic barriers
to unzipping of the DNA double helix. Proc Natl Acad Sci USA
98: 8608–8613.
Coury JE, Mcfail-Isom L, Williams LD and Bottomley LA (1996) A
novel assay for drug-DNA binding mode, affinity, and exclusion
number: scanning force microscopy. Proc Natl Acad Sci USA 93:
12,283–12,286.
Cuy Y and Bustamante C (2000) Pulling a single chromatin fiber
reveals the forces that maintained its higher order structure. Proc
Natl Acad Sci USA 97: 127–132.
Davenport RJ, Wuite GJL, Landick R and Bustamante C (2000)
Single-molecule study of transcriptional pausing and arrest by
E. coli RNA Polymerase. Science 287: 2497–2500.
Dammer U, Hegner M, Anselmetti D, Wagner P, Dreier M, Huber W
and Güntherodt HJ (1996) Specific antigen/antibody interactions
measured by force microscopy. Biophys J 70: 2437–2443.
Erie DA, Yang G, Schultz HC and Bustamante C (1994) DNA
bending by Cro protein in specific and nonspecific complexes:
implications for protein site recognition and specificity. Science
266: 1562–1566.
Essevaz-Roulet B, Bockelmann U and Heslot F (1997) Mechanical
separation of the complementary strands of DNA. Proc Natl Acad
Sci USA 94: 11,935–11,940.
Evans E and Ritchie K (1997) Dynamic strength of molecular adhesion
bonds. Biophys J 72: 1541–1555.
Evans E (1998) Energy landscapes of biomolecular adhesion and
receptor anchoring at interfaces explored with dynamic force
spectroscopy. Faraday Discuss 111: 1–16.
Evans E (2001) Probing the relation between force-lifetime-and
chemistry in single molecular bonds. Ann Rev Biophys Biomol
Struct 30: 105–128.
Flory PJ (1969) In: Statistical Mechanics of Chain Molecules, Interscience Publishers, New York.
Fotiadis D, Scheuring S, Muller SA, Engel A and Muller DJ (2002)
Imaging and manipulation of biological structures with the AFM.
Micron 33: 385–397.
Grange W, Husale S, Guntherodt HJ and Hegner M (2002) Optical
tweezers system measuring the change in light momentum flux. Rev
Sci Instrum 73: 2308–2316.
375
Guckenberger R, Heim M, Cevc G, Knapp HF, Wiegrabe W and
Hillebrand A (1994) Scanning-tunneling-microscopy of insulators
and biological specimens based on lateral conductivity of ultrathin
water films. Science 226: 1536–1538.
Guthold M, Zhu XS, Rivetti C, Yang GL, Thomson NH, Kasas S,
Hansma HG, Smith B, Hansma PK and Bustamante C (1999)
Direct observation of one-dimensional diffusion and transcription
by Escherichia coli RNA polymerase. Biophys J 77: 2284–2294.
Haber C and Wirtz D (2000) Magnetic tweezers for DNA micromanipulation. Rev Sci Instrum 71: 4561–4570.
Haijun Z, Yang Z and Zhong-Can O-Y (1999) Bending and basestacking interactions in double-stranded DNA. Phys Rev Lett 82:
4560–4563.
Hegner M, Wagner P and Semenza G (1993) Immobilizing DNA on
gold via thiol modification for atomic force microscopy. Imaging
in buffer solutions. FEBS Lett 336: 452–456.
Hegner M, Smith SB and Bustamante C (1999) Polymerization and
mechanical properties of single RecA-DNA filaments. Proc Natl
Acad Sci USA 96: 10,109–10,114.
Hegner M (2000) DNA handles for single molecule experiments. Single
Mol 1: 139–144.
Husale S, Grange W and Hegner M (2002) DNA mechanics affected
by small DNA interacting ligands. Single Mol 3: 91–96.
Jiao YK, Cherny DI, Heim G, Jovin TM and Schaffer TE (2001)
Dynamic interactions of p53 with DNA in solution by time-lapse
atomic force microscopy. J Mol Biol 314: 233–243.
Lebrun A and Lavery R (1996) Modelling extreme stretching of DNA.
Nucleic Acids Res 24: 2260–2267.
Leger JF, Robert J, Bourdieu L, Chatenay D and Marko JF (1998)
RecA binding to a single double-stranded DNA molecule: a
possible role of DNA conformational fluctuations. Proc Natl Acad
Sci USA 95: 12,295–12,299.
Leger JF, Romano G, Sakar A, Robert J, Bourdieu L, Chatenay D
and Marko JF (1999) Structural transitions of a twisted and
stretched DNA molecule. Phys Rev Lett 83: 1066–1069.
Li J, Cassell AM and Dai HJ (1999) Carbon nanotubes as AFM tips:
measuring DNA molecules at the liquid/solid interface. Surf Int
Anal 28: 8–11.
Lubensky DK and Nelson DR (2000) Pulling pinned polymers and
unzipping DNA. Phys Rev Lett 85: 1572–1575.
Marko JF and Siggia ED (1995) Stretching DNA. Macromolecules 28:
7016–7018.
Merkel R (2001) Force spectroscopy on single passive biomolecules
and single biomolecular bonds. Phys Rep 346: 343–385.
Odijk T (1995) Stiff chains and filaments under tension. Macromolecules 28: 7016–7018.
Otobe K and Ohani T (2001) Behavior of DNA fibers stretched by
precise meniscus motion control. Nucl Acid Res 29: U15–U20.
Perkins TT, Smith DE and Chu S (1997) Single polymer dynamics in
an elongational flow. Science 276: 2016–2021.
Rief M, Clausen-Schaumann H and Gaub HE (1999) Sequence
dependent mechanics of single DNA molecules. Nat Struct Biol 6:
346–349.
Rivetti C, Guthold M and Bustamante C (1996) Scanning force
microscopy of DNA deposited onto mica: equilibration versus
kinetic trapping studied by statistical polymer chain analysis. J
Mol Biol 264: 919–932.
Rivetti C, Guthold M and Bustamante C (1999) Wrapping of DNA
around the E. coli RNA polymerase open promoter complex.
EMBO J 18: 4464–4475.
Rivetti C, Walker C and Bustamante C (1998) Polymer chain statistics
and conformational analysis of DNA molecules with bends or
sections of different flexibility. J Mol Biol 280: 41–59.
Rivetti C and Codeluppi S (2001) Accurate length determination of
DNA molecules visualized by atomic force microscopy: evidence
for a partial B- to A-form transition on mica. Ultramicroscopy 87:
55–66.
Rocca AI and Cox MM (1997) RecA protein: structure, function, and
role in recombinational DNA repair. Prog Nucleic Acid Res Mol
Biol 56: 129–223.
Rouzina I and Bloomfield VA (2001a) Force-Induced Melting of the
DNA Double Helix. 1. Thermodynamic Analysis. Biophys J 80:
882–893.
Rouzina I and Bloomfield VA (2001b) Force-Induced Melting of the
DNA Double Helix. 2. Effect of Solution Conditions. Biophys J 80:
894–900.
Schäfer B, Gemeinhardt H, Uhl V and Greulich KO (2000) Single
molecule DNA restriction analysis in the light microscope. Single
Mol 1: 33–40.
Schumakovitch I, Grange W, Strunz T, Bertoncini P, Güntherodt H-J
and Hegner M (2002) Temperature dependence of unbinding
forces between complementary DNA strands. Biophys J 82: 517–
521.
Smith SB, Finzi L and Bustamante C (1992) Direct mechanical
measurements of the elasticity of single DNA molecules by using
magnetic beads. Science 258: 1122–1126.
Smith SB, Cui Y and Bustamante C (1996) Overstretching B-DNA: the
elastic response of individual double stranded and single stranded
DNA molecules. Science 271: 795–799.
Strick T, Allemand JF, Croquette V and Bensimon D (2000a) Twisting
and stretching single DNA molecules. Prog Biophys Mol Biol 74:
115–140.
Strick T, Allemand JF, Bensimon D and Croquette V (2000b) Stressinduced structural transitions in DNA and proteins. Ann Rev
Biophys Biomol Struct 29: 523–543.
Strick T, Croquette V and Bensimon D (2000c) Single-molecule
analysis of DNA uncoiling by type II topoisomerase. Nature 404:
901–904.
Strunz T, Oroszlan K, Schäfer R and Güntherodt HJ (1999) Dynamic
force spectroscopy of single DNA molecules. Proc Natl Acad Sci
USA 96: 11,277–11,284.
Strunz T, Oroszlan K, Schumakovitch I, Güntherodt H-J and Hegner
M (2000) Model energy landscapes and the force-induced dissociation of ligand-receptor bonds. Biophys J 79: 1206–1212.
Tanaka H, Hamai C and Kawai T (1999) High-resolution scanning
tunneling microscopy imaging of DNA molecules on Cu(111)
surfaces. Surf Sci 432: L611–L616.
Wenner JR, Williams MC, Rouzina I and Bloomfield VA (2002) Salt
dependence of the elasticity and overstretching transition of single
DNA molecules. Biophys J 82: 3160–3169.
Williams MC, Wenner JR, Rouzina I and Bloomfield VA (2001a)
Entropy and heat capacity of DNA melting from temperature
dependence of single molecule stretching. Biophys J 80: 1932–1939.
Williams MC, Wenner JR, Rouzina I and Bloomfield VA (2001b)
Effect of pH on the overstretching Transition of double-stranded
DNA: evidence of force-induced DNA melting. Biophys J 80: 874–
881.
Wuite GJL, Davenport RJ, Rappaport A and Bustamante C (2000)
An integrated laser trap/flow control video microscope for the
study of single biomolecules. Biophys J 79: 1155–1167.